What starts as "fix the loan flow" quickly unravels into a system where every stakeholder has already built a better workaround than your product. Vendors on WhatsApp. Counsellors on spreadsheets. Students going directly to banks. The real brief wasn't a redesign. It was earning back adoption from people who had already moved on.

01 - Problem

The infrastructure existed. Almost none of it was being used.

Edvoy Funds wasn't a new idea it was a failing one. The service existed to help international students secure education loans, and on paper, the pieces were in place: vendor partnerships, a counsellor team, even a vendor portal. In practice, almost none of it was running through the product.

Student loan requests were being collected manually. Counsellors were submitting details over WhatsApp and email. Vendors were receiving student leads, processing loans, and in some cases reporting nothing back, because there was no system in place to verify what had actually happened. Commission tracking lived in spreadsheets. Document trails were scattered across chat threads.

The most damaging pattern wasn't inefficiency. It was the accountability gap. Vendors could and did close loans that originated from Edvoy and claim the commission themselves, because there was no infrastructure to dispute it. Revenue was walking out the door, and the business had no reliable way to know how much or where.

What got lost:

What was approved · What was disbursed · What commission was owed · Who did what

The cost of that:

Vendors received Edvoy leads, closed loans, reported nothing back. No audit trail. No way to dispute. Revenue walked out the door and the business had no reliable way to know how much.

So we asked:

How do we build a system valuable enough that people willingly move their work back into it?

02- Research

The problem underneath the problem

I needed to understand how the financing ecosystem actually operated, not how it was supposed to.

Every workaround was rational. That's what made this hard.

The instinct when a product isn't being used is to ask: what's wrong with the UI? But that's the wrong question when adoption failure is a value failure, not a usability failure.

I spent time with all three stakeholder groups before touching a single screen.

Students

Students told me they often didn't know Edvoy Funds existed at all. The ones who did know said external loan agents felt faster, clearer, more responsive — so they chose them. Edvoy wasn't losing students to competitors. It was losing them to general disorganisation and late discovery. Most students only started thinking about financing after they received a university offer letter, which meant the window was already compressed by the time Edvoy entered the picture.

Counsellors

Counsellors revealed a different problem. The financing ecosystem had nearly ten vendor partnerships, each with different commission structures, regional coverage, eligibility criteria, and negotiated interest concessions. A handful of senior counsellors knew all of this. Most didn't — and because none of it lived in a system, knowledge transfer depended entirely on who happened to be available and willing to share. Students got inconsistent guidance not because counsellors didn't care, but because the institution was scaling faster than its own operational knowledge.

Vendors

The vendor interviews were the most clarifying. I went in expecting usability complaints. What I found was simpler and harder to fix: the portal didn't reduce their workload, so why would they use it? WhatsApp is faster. Email has search. The portal had neither advantage and added a step without removing one. That's not a design failure. That's a value failure — and the only fix is to make the product more useful than the workaround already in place.

Every workaround I found was rational. The WhatsApp groups, the spreadsheets, the vendor ignoring the portal — none of it was laziness. The system had stopped being worth the friction of using, and people built better alternatives and moved on. That framing changed everything about how we approached the rebuild.

"The portal adds a step. It doesn't remove one."

It wasn't a design failure. It was a value failure. And you can't fix a value failure with better UI.

03 - The real scope

The loan journey starts with an offer letter. So did we.

Once a student received a university offer, they could apply for a loan directly in the Edvoy app. Agents could submit on their behalf — but only after the student gave explicit consent. Accountability built in from step one.

Regional routing

10 vendors. Not all relevant to every student. A student applying from Chennai gets routed to the 3–4 vendors who cover South India. Not all 10. Reduced noise. More relevant leads. Vendors taken more seriously.

Adaptive document collection

Non-collateral loans need fewer documents. Collateral loans need more. The platform knew the difference and asked only for what was relevant.

More importantly - everything was centralised and timestamped. "I already sent that" vs "we don't have it" now had a definitive answer. A full category of daily friction, eliminated.

Vendor portal - rebuilt around their workflow

Old portal: a reporting layer built for Edvoy's benefit. New portal: an operational tool built around what vendors actually do.

Access student docs → Upload approval letters → Upload disbursal records → Track progress.

That's it. When the portal reduced work instead of adding it, they used it.

OCR-powered processing

Vendors upload approval letters. OCR extracts loan values, disbursal details, lender info. Forms pre-filled for Funds Advisors to review.

Manual entry dropped. Inconsistencies dropped with it.

Notes and tasks, the WhatsApp replacement

Notes and tasks already existed in the platform. Nobody used them, same reason nobody used the vendor portal.

We rebuilt them to be integrated, contextual, and genuinely useful. Tag a student. Tag an application stage. Assign to a counsellor. Attach documents. Set deadlines. Track resolution.

Counsellors stopped DMing each other on WhatsApp because the task was already on the student profile. Funds Advisors stopped forwarding documents over chat because they could attach them to the relevant stage.

Communication moved into the system because the system finally held the context. This doesn't look impressive in a screenshot. It's a compliance win, an audit win, and an accountability win.

The commission engine

The most complex part of the project. The most direct business impact.

Why loan commissions are different:

Commission isn't based on the approved loan amount. It's based on what's actually disbursed.

A student approved for ₹50 Lakhs → draws ₹15 Lakhs in year one. Commission: calculated on ₹15 Lakhs only. Remaining ₹35 Lakhs → disbursed in parts, over time.

That means:

One student · Multiple disbursals · Multiple invoices · Multiple commission calculations · Ongoing tracking

The database had to be rethought entirely, mapping multiple disbursals and invoices back to a single application, a single student, a single vendor relationship.

What we built:

A commission engine that automated the entire cycle.

Disbursal verified → Commission calculated (vendor split + agent split + tax) → Invoice auto-generated → Sent to vendor → Payment tracked

If the disbursed amount is less than approved: loop stays open. Closes only when the full approved amount is drawn down or the application is resolved.

Finance teams got real-time revenue visibility for the first time. Leadership could see, live, what was being earned and what was projected.

What we didn't build

Loan recommendation engine

Match students to the most suitable lender based on profile. Surface applicable concessions. Flag Edvoy's own 2% interest benefit. Right idea. Wrong timing. Building intelligence on a broken foundation amplifies chaos. Infrastructure first.

Loan repayment via Edvoy app

Flywire integration. Students manage repayments in-app. Transaction revenue on top of commissions. Strong product idea. Strong commercial idea. Still waiting.

Knowing what to cut — and being clear about why — is part of the job. These weren't compromises. They were sequencing decisions.

04 - Outcomes

The platform launched across six surfaces: Edvoy App, Student Portal, CRM, Edge, Vendor Portal, and Cloud Admin.

₹70 Lakhs profit in the first intake post-relaunch
1–2 weeks → 1–2 days loan processing timelines
48 hours express approval turnaround
6 platforms rebuilt end-to-end

Everything visible. Nothing off the table. Commissions tracked. Vendors accountable. Students informed.

But the reaction that landed hardest internally wasn't about the student experience improvements. It was the real-time commission tracking and revenue forecasting.

For the first time, the business could see what was being earned, what was pending, and what was projected - without chasing anyone for a status update. That visibility closed the accountability loop that had been leaking revenue for as long as the service had existed.

05- Reflection

At some point in this career you stop asking "how should this look" and start asking "what does this need to replace."

The hardest design problem on this project wasn't the commission engine or the database restructuring or the six-platform scope. It was accepting early on that better UI wasn't the answer. The workarounds people had built were genuinely good. WhatsApp is fast. Spreadsheets are flexible. You can't out-design tools people already trust unless you offer them something those tools can't.

The shift that unlocked the project was moving from "how do we make the portal better" to "what does this portal need to be worth — to a vendor who already has a faster option?" That question changes the entire design brief. It moves you from optimising an interface to rethinking the value exchange.

The commission engine is the part I'm most proud of technically — the complexity of mapping multi-disbursal loans to automated invoice generation across vendor and agent splits is genuinely hard. But the part that made the most difference was simpler: giving counsellors loan status visibility inside the tools they already lived in, so they didn't have to maintain a parallel mental model of where each student's application was.

Nine years of this work has made me increasingly suspicious of solutions that require behaviour change as a precondition. The best products I've worked on don't ask people to change how they work — they make the new way easier than the old way. Edvoy Funds, at its core, is a case study in earning adoption back from workarounds that had every reason to exist.

The ₹70 Lakhs is the number. The thing I'm more proud of is that nothing is happening off the table anymore.

Tags: Systems rebuild · Multi-stakeholder UX · Commission engine · Finance automation · OCR integration · B2B + B2C · Operational accountability · Value-driven adoption